Leaf area determinations in sorghum and maize by the length-width method.

Authors

  • H. van Arkel

DOI:

https://doi.org/10.18174/njas.v26i2.17096

Abstract

5 plant population trials were carried out in 1974 and 1975 with 5 sorghum and 4 maize cv. Each trial was sample harvested several times during the growing period. In 6080 leaves the leaf area was carefully determined and related to its length and width. By regression calculations an examination was made regarding the construction of the best calibration lines for the estimation of the product of length and width from leaf area. Plant population had no significant effect on the regressions. Drawing the regression lines through the origin was sometimes statistically rejected but if the regression lines were forced through the origin in spite of this, the loss of precision was very small. It is therefore concluded that forcing the regression lines through the origin involves an acceptable estimation error. The resulting regression lines which were established separately for each cv. and harvest proved to estimate the leaf area accurately. The values of the regression coefficients varied between 0.65 and 0.77. The lowest correlation coefficient was 0.968, but 55% of all correlation coefficients were >0.990. If data from different harvests were pooled (although not allowable statistically) the highest mean error of estimation of leaf area at any harvesting date was 8.5%. (Abstract retrieved from CAB Abstracts by CABI’s permission)

Downloads

Published

1978-05-01

Issue

Section

Papers